ConvoMap: Using Convolution to Order Boolean Data
Heatmaps, also called matrix visualisations, are a popular technique for visualising boolean data. They are easy to understand, and provide a relatively loss-free image of a given dataset. However, they are also highly dependent on the order of rows and columns chosen. We propose a novel technique, called ConvoMap, for ordering the rows and columns of a matrix such that the resulting image represents data faithfully. ConvoMap uses a novel optimisation criterion based on convolution to obtain a good column and row order. While in this paper we focus on the creation of images for exploratory data analysis in binary data, the simplicity of the ConvoMap optimisation criterion could allow for the creation of images for many other types of data as well.
- 2.Bollen, T., Leurquin, G.: Faithful visualization of categorical data. Master’s thesis (2017). Université catholique de Louvain, Louvain-la-NeuveGoogle Scholar
- 5.Hahsler, M., Hornik, K. Buchta, C.: Getting things in order: an introduction to the R package seriation. J. Stat. Softw. Artic. 25(3) (2008)Google Scholar
- 7.Kaski, P., Junttila, E.: Segmented nestedness in binary data. In: Proceedings of the Eleventh SIAM International Conference on Data Mining, pp. 235–246 (2011)Google Scholar
- 9.Junttila, E.: Patterns in permuted binary matrices. PhD thesis, University of Helsinki (2011)Google Scholar
- 11.Christofides, N.: Worst-case analysis of a new heuristic for the travelling salesman problem, Report 388, Graduate School of Industrial Administration, CMU (1976)Google Scholar